Wirth's Table-driven Parser

"Daniel Zingaro" <zingard@mcmaster.ca>
15 May 2006 22:23:34 -0400

          From comp.compilers
Related articles
Wirth's Table-driven Parser zingard@mcmaster.ca (Daniel Zingaro) (2006-05-15)
Re: Wirth's Table-driven Parser Juergen.Kahrs@vr-web.de (=?UTF-8?B?SsO8cmdlbiBLYWhycw==?=) (2006-05-16)
| List of all articles for this month |
From: "Daniel Zingaro" <zingard@mcmaster.ca>
Newsgroups: comp.compilers
Date: 15 May 2006 22:23:34 -0400
Organization: Compilers Central
Keywords: history, question
Posted-Date: 15 May 2006 22:23:34 EDT

Hi all,


Out of interest's sake:


In sec. 4.3 of Wirth's 1996 Compiler Construction book, he gives a procedure
called Parsed for parsing a graph representing a kind of parse table.
Without modifying Parsed, is it possible to cope with nested EBNF
structures, for example


x = a {[y] z} b


?


I feel that the treatment of the empty nodes makes it impossible to parse
sentences of this grammar correctly. I know Wirth has a solution in the
original (non-english) manuscript: if anyone has that solution, or other
insight, I would appreciate it.


Thanks,
Dan


Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.